Offensive Security RL Researcher

idler•San Francisco, CA

1d•$200,000 - $500,000•Onsite

About The Position

idler is a frontier data research lab that builds evaluation environments and datasets for leading AI labs. After raising a $9m seed round, idler has developed coding evaluations for top coding models and expanded into other domains including cybersecurity. The company is experiencing high demand for its data and is scaling its team. The ideal candidate will lead idler's cybersecurity efforts, including defining the roadmap for environments, building and running the systems to produce them, managing part-time experts, and engaging with researchers at frontier AI labs to understand their needs and sell them solutions. The role requires a strong security researcher who can also develop software to implement their research. If idler's investment in cybersecurity changes, the role may transition to another project within the company.

Requirements

3+ years of experience in a professional cybersecurity role, with a track record in the top 5% of peers by reputation.
Deep hands-on experience in at least one of: web app pentesting, bug bounty hunting, vulnerability/zero-day research, hacking agents, or AI security.
Well-connected within the security community.
Comfortable writing and shipping production code (Python, TypeScript, or similar).
Strong written communication skills.

Nice To Haves

A public bug bounty track record, live-hacking-event invitations, Pwn2Own placements, or published original research.
Experience managing or mentoring other security researchers or hackers.
Experience thinking about how to measure or grade security work beyond pass/fail.
Experience building tools or automation to support research.

Responsibilities

Drive the roadmap for idler's cybersecurity evals and environments, and decide which environments to build next and where to allocate engineering resources.
Build and run the systems that produce those environments, writing production code.
Recruit, onboard, and manage part-time cybersecurity experts to review environments for fairness, correctness, and quality.
Work directly with researchers at frontier AI labs to understand their cybersecurity capability needs and sell them environments that meet those needs.